import pandas as pd 
import numpy as np 
import seaborn as sns
from matplotlib import pyplot as plt
%matplotlib inline
import glob
import os
import warnings; warnings.simplefilter('ignore')

print("Current Directory:", os.getcwd())
list_of_csv_logs = glob.glob("*.csv")
print(list_of_csv_logs)

Current Directory: /Users/angelateng/Documents/Github/supsup/csv_logs
['slurm_supsup_basis_gpu_randmask_3_results.csv', 'slurm_supsup_basis_gpu_randmask_15_results.csv', 'slurm_supsup_basis_gpu_randmask_10_results.csv', 'slurm_supsup_basis_gpu_20_results.csv', 'slurm_supsup_basis_gpu_12_results.csv', 'slurm_supsup_basis_gpu_5_results.csv', 'slurm_supsup_basis_gpu_randmask_5_results.csv', 'slurm_supsup_basis_gpu_optimal_start_10_results.csv', 'slurm_supsup_basis_gpu_optimal_start_7_results.csv', 'slurm_supsup_basis_gpu_7_hyperparams_64_8e-3_50_results.csv', 'slurm_supsup_basis_gpu_3_results.csv', 'slurm_supsup_basis_gpu_randmask_7_results.csv', 'slurm_supsup_basis_gpu_optimal_start_5_results.csv', 'slurm_supsup_basis_gpu_optimal_start_12_results.csv', 'slurm_supsup_basis_gpu_7_hyperparams_64_1e-3_50_results.csv', 'slurm_supsup_basis_gpu_7_hyperparams_64_5e-3_50_results.csv', 'slurm_supsup_basis_gpu_7_hyperparams_64_1e-2_50_results.csv', 'slurm_supsup_basis_gpu_15_results.csv', 'slurm_supsup_basis_gpu_randmask_12_results.csv', 'slurm_supsup_basis_gpu_10_results.csv', 'slurm_supsup_basis_gpu_randmask_20_results.csv', 'slurm_supsup_basis_gpu_7_hyperparams_64_3e-2_50_results.csv', 'slurm_supsup_basis_gpu_7_results.csv']

for item_glob in list_of_csv_logs : 
    current_df = pd.read_csv(item_glob)
    
    sns.relplot(x="sparsity", y="accuracy", hue = "task", legend="full",  ci=None, kind="line", data=current_df)
    plt.title(str(item_glob) + ' Performance (Accuracy)')
    sns.relplot(x="epochno", y="accuracy", hue = "task", legend="full",  ci=None, kind="line", data=current_df)
    plt.title(str(item_glob) + ' Performance (Accuracy)')

    sns.relplot(x="sparsity", y="train_loss", hue = "task", legend="full",  ci=None, kind="line", data=current_df)
    plt.title(str(item_glob) + ' Performance (Train Loss)')
    sns.relplot(x="epochno", y="train_loss", hue = "task", legend="full",  ci=None, kind="line", data=current_df)
    plt.title(str(item_glob) + ' Performance (Train Loss)')

    sns.relplot(x="sparsity", y="test_loss", hue = "task", legend="full",  ci=None, kind="line", data=current_df)
    plt.title(str(item_glob) + ' Performance (Test Loss)')
    sns.relplot(x="epochno", y="test_loss", hue = "task", legend="full",  ci=None, kind="line", data=current_df)
    plt.title(str(item_glob) + ' Performance (Test Loss)')